Halden your writings in black, my comments in green

The peak current amplitude fell a little bit?????? (750-650 mApp)? but the shoulder amplitude rose substantially (500 vs 200 mApp).? --yes-

? The phase lag dropped from about 2ms to 1.5 ms.? –--yes- about expected-

The current waveform peak concurs with the voltage crossing, so this doesn't represent a difference in dissipated power.?–Yes quite close?? -

?But at the time of the voltage peak, the current is about 2.5 times what it was without the short. ---yes expected-

To estimate the power dissipation represented in the second image (#20), I ignored the peak that occurs when no current flows and approximated the rest of the waveform as a sinusoid with amplitude 540 mApp, 190 mA rms.?????? Multiplying by 113 Vrms and cos(30 deg) gives 19 W of power dissipation.

If spread throughout the transformer, this would cause the transformer to get noticeably warm.? Well Halden I wonder if all this watts stuff really matters, but in my way of thinking #16 does indicate power too, likely in the core and some in the primary turns resistance; I’m guessing that it is mostly core loss since we see it running into some saturation, but we can’t be sure.? As I see it, in #20 the total heat is the heat from #16 plus the heat going into the shorted area.

So the shorted turn area takes 540 mApp? minus about 200 mApp looking at Vmax which still means that the short is taking a considerable and dominant current and… power [since losses in the primary must be very low at this current]. I stole your images and generated this

At the lower part you can see the difference, which shows pretty much what you said but one can see the waveform

If I did it right you can see the difference [shorted turn] is somewhat sinusoidal

?

But its probably mostly dissipated at the site of the short—yyy ??? - or in the shorted winding itself, --yyy-? either of which would get quite hot after some time.?–yyy mostly-? well, I need to revisit this transformer stuff, especially for this rather lopsided multi-winding unit. Stating some obvious things; if there is a high resistance there will be little heat generated at the short or in the turn, but if the short is solid? there will be lots of heat in the turn and none at the short, and depending mostly on the resistance of the wire the worst case might be when the ohms-[short]? = about ohms-[1 turn] and so about half the heat will be in the short and half in the winding turn, and in time all nearby insulation will deteriorate until doomsday comes, all to say it might start at the point of the T-T short and migrate mostly to the Whole turn. That becomes interesting because the longer the heat is generated at the point the more likely that local area will really overheat and destroy insulation

?

That heat might damage adjacent windings and possibly cause additional shorts and progressing failures until a fuse blows, interrupting the process. Yyy

Halden ?good that the HV T-T seems ok to your eye.

To my eye #16 looks like magnetizing current only. I have wondered about letting it run for some hours while your nose is not to far away.

One thing I have wanted is a thermal camera; it might be useful here. ????

?

Halden, ? ?

revised? oops this very draft post somehow flew off the table ?too soon, I never learn? !?

some parts were to be revised or deleted, and others added

I am leaving the original as just sent? except for this oops note.

?

I was/am surprised that the primary current shown in the img 20 test is so low. The first thought is that an alligator or even crock might not have the pressure/force to make a good “electrical connection joint”; ?especially at say 30 Amps.

Just to get a ball park idea of the resistance , I made a real guess that the 5volt wdg has 1 foot of #16 per turn = 0.004 ohms.

I think you mentioned .5 volts per turn for this transformer? so? .5/.004 ??=125 amps. I should rethink that guess-work but ?for starters a, simple light pressure contact of round wire to round wire using an alligator clip will probably develop too much resistance. Just as an example, advertising indicates 5 to 10 milli-Ohms for a brand new alligator clip [I stored some examples in my docs].?

But if indeed your shorted turn test on this transformer design is true, I will have to rethink my thoughts about failures in the secondary’s. Since this T-T failure “load” will allow the transformer to cook internally until the heat in that spot creates another fault that manages to blow the fuse, of maybe set-off a smoke detector etc.

??You might say this transformer is four transformers in one, and the 5 volt VA winding is a small fraction of the primary, so when it comes to the 5V scdy’s [rated at 15VA by Hammond] a solid T-T short ?may well roast the whole area to carbon while the primary [somewhere about 300 VA] sees only a smallish current, because of the small scdy wire

?So… ???

?

Looking at the results so far, I wonder if ?these shorted turn tests will be of much value. It is good to experiment and see what happens, but a T-T short is a well known killer, and a transformer with a T-T will be useless sooner or later. ?All you can hope to do is establish that you are seeing normal magnetizing current.

Your img 16 looks like typical magnetizing current only, but you can never be sure that there is nothing else.

?

At this point, it seems to me that what is wanted/needed is insurance that the HV winding has no T-T failure at “that” spot, but THAT is a major challenge, even for the industry. However if it has failed badly, a HV T-T should resemble this 5v ?alligatorized T-T test {img 20]. Since the primary and secondary has R and an unknown X , power/heating calculations would be beyond me; but I would still look at the current. ??

?

Hovever, a real good ?magnifier looking at that spot might reveal if the previous arcing “mucked-up” the enamel between turns, as mentioned long ago by somebody.? And Halden, I am thinking that the big hole in the insulation is due to corona’s ozone and the arc? diameter was thin

?

?

Halden, ?Jim, ?now a bit late.. ???

Measuring values on a scope can easily return a peak value, and depending on how you measure, P-P might be more accurate. I certainly never managed to see an RMS or Average value on a scope screen; except for pure DC. And it is tough to read an accurate value at each microsecond, and calculate RMS, but I guess it can be done. ?I use which-ever is available at the time; and does the wanted job . ?

?

Halden, I will bypass most of your writings for now and just say the first? img {16} looks as expected.

The second img {20} with the shorted turn has a current waveform as expected, but it is strange that the current amplitude is so low. Just to be sure is the voltage measured at the transformer primary? And not into the dim-bulb? ?I am missing something.